DOE/NETL-2001/1158

Advanced Flue Gas Desulfurization (AFGD)Demonstration ProjectA DOE Assessment

August 2001

U.S. Department of EnergyNational Energy Technology Laboratory

P.O. Box 880, 3610 Collins Ferry RoadMorgantown, WV 26507-0880andP.O. Box 10940, 626 Cochrans Mill RoadPittsburgh, PA 15236-0940

website: www.netl.doe.gov

2

Disclaimer

This report was prepared as an account of work sponsored by anagency of the United States Government. Neither the United StatesGovernment nor any agency thereof, nor any of their employees, makesany warranty, express or implied, or assumes any legal liability orresponsibility for the accuracy, completeness, or usefulness of anyinformation, apparatus, product, or process disclosed, or representsthat its use would not infringe privately owned rights. Referencetherein to any specific commercial product, process, or service by tradename, trademark, manufacturer, or otherwise does not necessarilyconstitute or imply its endorsement, recommendation, or favoring bythe United States Government or any agency thereof. The views andopinions of authors expressed therein do not necessarily state or reflectthose of the United States Government or any agency thereof.

3

Contents

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

I Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

II Technical and Environmental Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

II.A Promise of the Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

II.A.1 Conventional FGD Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 II.A.2 AFGD Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

II.B Process Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9II.C Project Objectives/Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11II.D Environmental Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12II.E Post-Demonstration Achievements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

III Operating Capabilities Demonstrated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

III.A Size of Unit Demonstrated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13III.B Performance Level Demonstrated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13III.C Major Operating and Design Variables Studied . . . . . . . . . . . . . . . . . . . . . . . . 14III.D Boiler Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16III.E Commercialization of the Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

III.E.1 Current Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 III.E.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

IV Market Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

IV.A Potential Markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18IV.B Economic Assessment of Utility Boiler Applications . . . . . . . . . . . . . . . . . . . . . 18

IV.B.1 AFGD Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 IV.B.2 Comparison With Other Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . 19

V Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

4

List of Figures and Tables

Figure Page

1 Pure Air Advanced Flue Gas Desulfurization Process . . . . . . . . . . . . . 9

2 Absorber Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

3 Effect of Slurry Circulation on SO2 Removal Efficiency . . . . . . . . . . . . 14

4 Effect of Stoichiometric Ratio on SO2 Removal Efficiency . . . . . . . . . . 15

5 Effect of Liquid/Gas Ratio on SO2 Removal Efficiency . . . . . . . . . . . . . 15

Table Page

1 Coal Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2 Summary of Performance and Cost Data . . . . . . . . . . . . . . . . . . . . . . . 20

5

Executive Summary

This document serves as a U.S. Department of Energy (DOE) post-projectassessment of the Clean Coal Technology (CCT) Round II Advanced Flue GasDesulfurization (AFGD) Demonstration Project, conducted by Pure Air. Pure Airis a general partnership between Air Products and Chemicals, Inc., and MitsubishiHeavy Industries America, Inc. In December 1989, Pure Air entered into anagreement to conduct this study, with Northern Indiana Public Service Company(NIPSCO) as the host and cosponsor. DOE provided 42 percent of the totalproject funding cost of $152 million. The demonstration operations were con-ducted from June 1992 to June 1995 at NIPSCO’s Bailly Generating Station (units7 and 8) located in Chesterton, Indiana, to treat the combined flue gases from twoboilers with a total nameplate capacity of 616 MWe.

The AFGD process accomplishes sulfur dioxide (SO2) removal in a single absorberwhich performs three functions: prequenching, absorption of SO2, and oxidation toproduce gypsum. The performance objectives of this project were to

& Remove at least 90-percent SO2, with a target of 95 percent.& Reduce process cost by one-half that of conventional flue gas

desulfurization (FGD).& Reduce space requirements.& Produce wallboard-grade gypsum.

These performance objectives were met except for process cost which wasreduced to 63 percent of conventional FGD cost. The SO2 removal target wasexceeded. For the five midwestern bituminous coals tested, with sulfur contentsranging from 2.21 to 4.73 wt%, SO 2 removal efficiency averaged 94 percent,with a maximum of over 98 percent. The demonstration facility was operated forabout 26,300 hours, with system availability of 99.5 percent. Over 210,000 tons ofwallboard-grade gypsum were produced, having an average purity of 97.2 percent.

Costs were estimated for a 500-MWe AFGD unit, using a projected processdesign which incorporates improvements based on experience gained from thedemonstration project. The coal feed is assumed to contain 3 wt% sulfur, and SO 2

emissions are assumed to be reduced by 90 percent. The capital cost is $111/kW.For a 15-year project life, the levelized cost on a current-dollar basis is 5.3mills/kWh, which is equivalent to $245/ton of SO2 removed. The levelized cost forAFGD is about 63 percent of that for conventional wet limestone desulfurization.This is a significant cost reduction, approaching the target value of 50 percent.Space requirements for AFGD are substantially lower than those for conventionalFGD processes.

6

The project received two major awards: the Outstanding Engineering Achievementaward, from the National Society of Professional Engineers in 1992; and thePowerplant of the Year award from Power magazine in 1993, for demonstratingadvanced limestone FGD technology with innovations in wastewater treatment andgypsum production.

The AFGD unit remains in operation at the Bailly Station, where it is performingvery well. With increasingly stringent air quality regulations, AFGD technologyshould be a major contender in a growing market for flue gas cleanup. In addition,the innovative use of gypsum by-product in wallboard manufacture has establisheda new trend; synthetic gypsum produced at FGD facilities has become thepreferred feedstock for wallboard manufacture because its uniform propertiessimplify manufacturing operations for existing users .

7

I Introduction

The goal of the U.S. Department of Energy (DOE) Clean Coal Technology (CCT)program is to furnish the energy marketplace with a number of advanced, moreefficient, and environmentally responsible coal utilization technologies throughdemonstration projects. These projects seek to establish the commercial feasibilityof the most promising advanced coal technologies that have developed beyond theproof-of-concept stage.

This document serves as a DOE post-project assessment of the of the CCT Round II Advanced Flue Gas Desulfurization (AFGD) Demonstration Project, asdescribed in a Report to Congress (U.S. Department of Energy 1989). InDecember 1989, Pure Air and Northern Indiana Public Service Company(NIPSCO) entered into a cooperative agreement to conduct the study. Pure Air isa general partnership between Air Products and Chemicals, Inc., and MitsubishiHeavy Industries America, Inc. Subsequently, Pure Air on the Lake, L.P., wasorganized as a project company of Pure Air to implement the project. NIPSCOserved as the host at its Bailly Generating Station. DOE provided 42 percent of thetotal project cost of $152 million.

The demonstration operations were started in June 1992 and completed in June1995. The independent evaluation contained herein is based primarily oninformation from the project Final Report (Pure Air 1996), as well as sourceslisted in the bibliography.

The AFGD process removes sulfur dioxide (SO2) in a single absorber whichperforms three functions: prequenching, absorption, and oxidation to wallboardgrade gypsum.

The Clean Air Act, enacted in 1970 and amended in 1977, established New SourcePerformance Standards (NSPS) for emissions of SO2, nitrogen oxides (NOX), andparticulates from stationary coal-fired power plants. These regulations were mademore stringent in the Clean Air Act Amendments (CAAA) of 1990.

The host site chosen for this CCT demonstration project, NIPSCO’s BaillyGenerating Station, is located along the shore of Lake Michigan approximately 12miles northeast of Gary, Indiana. The station is bordered by industrial installationsand by the Indiana Dunes National Lakeshore.

The performance objectives of this project were to& Remove at least 90-percent SO2, with a target of 95 percent.& Reduce process cost by one-half that of conventional flue gas

desulfurization (FGD).& Reduce space requirements.& Produce wallboard-grade gypsum.

8

II Technical and Environmental Assessment

II.A Promise of the Technology

II.A.I Conventional FGD Processes

Conventional wet-process systems for flue gas desulfurization use an aqueousslurry of limestone (CaCO3) as the reagent in an absorber, or scrubber, which isusually a vertical vessel in which the flue gas is contacted with the slurry. The flowof gas, which is normally countercurrent to the liquid, is limited by the masstransfer characteristics of the system, thereby determining the cross sectional areaof the scrubber. These vessels tend to be quite large in practice. The reactionproduct exiting the bottom of the scrubber is an aqueous sludge containing calciumsulfite (CaSO3) and calcium sulfate (CaSO4), which is sent to a disposal pond orlandfill after partial dewatering. Such ponds require large areas of land and areunsightly.

In addition to the disposal problem, another disadvantage of conventionalprocesses is potential scaling and plugging of the scrubber and auxiliary pipingresulting from the presence of these two calcium salts. This in turn requiresfrequent shutdowns for maintenance and results in the need for a spare scrubbermodule of equivalent capacity to permit uninterrupted treatment of the flue gaswhile the main scrubber is off line. Alternative processes have been developed inwhich the sludge is reacted with oxygen to convert the CaSO3 to CaSO4 (gypsum),which can be sold as wallboard or used in cement manufacture. Generally this steprequires a separate reaction vessel, using air as the oxidizing agent.

II.A.2 AFGD Process

The Pure Air project was undertaken to evaluate the technical and economicfeasibility of using the AFGD process to remove SO2 from the flue gas of a coal-fired boiler in a single, highly efficient contacting device which providesprequenching of the flue gas, absorption, and oxidation. Pulverized limestone isinjected directly into the absorber. Oxidation is achieved by means of an air rotarysparger (ARS), which provides sufficient agitation and air distribution to achieveessentially complete oxidation of sulfite. The process incorporates a uniqueagglomeration step to enhance the physical properties of the gypsum by-product.The resulting product, called PowerChip® gypsum, is easier for existing users tohandle in their wallboard manufacturing operations. In addition, the AFGD systemincludes a wastewater evaporation system (WES), which eliminates wastewaterdischarge by evaporating the residual water using heat contained in the flue gas.

9

II.B Process Description

A schematic flowsheet showing the AFGD process is given in Figure 1. Theprocess involves reaction of SO2 with water to form sulfurous acid (H2SO3), whichin turn is oxidized to sulfuric acid (H2SO4). The latter reacts with limestone toform gypsum. The primary equations are as follows:

SO2 + H2O YY H2SO3

H2SO3 + ½ O2 YY H2SO4

CaCO3 + H2SO4 + H2O YY CaSO4 . 2H2O + CO2

Alternatively, the reaction path can be described as follows, with the same netresult:

SO2 + H2O YY H2SO3

CaCO3 + H2SO3 YY CaSO3 + CO2 + H2O

CaSO3 + ½ O2 + 2H2O YY CaSO4 . 2H2O

Figure 1. Pure Air Advanced Flue Gas Desulfurization Process

The Pure Air facility utilizes a single resin-lined absorber or “scrubber” module totreat all of the flue gas from the Bailly Station’s two coal-fired boilers. There is nobackup or spare scrubber. The absorber operates with cocurrent flow of flue gasand scrubbing slurry, with two levels of slurry distribution. The absorber designincludes a large gas-liquid disengagement zone, which is conducive to a relatively

10

W et/D ryIn terface

W as h Header

A bs orberRe circu lation

He aders

A bs orber G rid

A bs orberTank Ag ita tors

A ir Ro ta ryS parg er (A RS )

A bs orberTank

A bs orberFee d Pum ps

C leanedF lue G asLeav ing

A bs orber

M istE lim ina tor

F lue G asE ntering A bsorber

2K-1069-2 C

high flue gas flow-rate of about 20 ft/sec, resulting in a compact absorber design.By using high-efficiency open-grid packing, tower height is also reduced.

The absorber performs three separate functions in the same vessel: prequenchingthe flue gas, absorption, and oxidation. In older FGD systems, these functionsrequired separate vessels. Additional space- and cost-saving features include

& A non-pressurized slurry distribution system, requiring approximately 30 percent less recirculation-pump power than conventional countercurrent spray towers.

& Fountain-like flow that does not generate a fine mist, thereby reducing mist eliminator loading by as much as 95 percent compared to countercurrent designs.

& Use of a dry pulverized limestone injection system, eliminating the need for ball mills, tanks, pumps, and other equipment associated with on-site wet grinding systems.

The absorber includes the ARS concept, which combines the functions of mixingand air distribution. There are two ARSs installed in the absorber at Bailly Station.In conventional FGD systems with forced oxidation, mixing is done by agitators inthe scrubber while oxidation takes place in a separate vessel with a fixed airsparger. The ARS system provides higher oxygen utilization, improved mixing,lower agitation power and reduced maintenance. A schematic diagram of theabsorber is given in Figure 2.

Figure 2. Absorber Module

Raw gypsum slurry is pumped to a batch centrifuge, where water is removed andthe cake is washed to recover wallboard-grade gypsum. The major portion of thefiltrate is recycled to the absorber. The net liquid effluent is treated in the WES,which involves injection into the flue gas duct upstream of the existing electrostatic

11

precipitator (ESP). The hot flue gas evaporates the water and the dissolved solidsare collected in the ESP along with the fly ash.

The gypsum by-product from the AFGD process has the consistency of wet sand,which is not conducive to transportation or handling by existing wallboardmanufacturing equipment. The CCT demonstration project included thedevelopment of the proprietary PowerChip® gypsum agglomeration process. Thisprocess utilizes a compression mill, operating at a unique combination oftemperature and pressure, to reformulate and modify the physical structure of thegypsum to produce stable, semi-dry agglomerated flakes resembling the propertiesof natural gypsum. The resulting by-product can be transported and handled aseasily as the natural material. PowerChip® gypsum is produced in relatively dry,consistently sized chips which do not freeze together in cold weather.

Gypsum made at the Bailly Station is sold to United States Gypsum Company(U.S. Gypsum) for wallboard manufacture at its East Chicago, Indiana plant. Thisplant was the first facility in North America to produce wallboard from 100-percent FGD gypsum. Since production began, U.S. Gypsum has taken the entireoutput of the Bailly AFGD unit. During the 3-year demonstration, gypsumproduction exceeded 210,000 tons. Because there is added cost associated withrunning PowerChip® gypsum, and because U.S. Gypsum is capable of handlingthe normal production of gypsum from Bailly Station, only a portion of the by-product was converted to PowerChip® gypsum.

Another feature of this project is a novel business concept whereby Pure Air is theowner of the AFGD unit, operating the system for the utility under a servicecontract. Under this agreement, Pure Air is responsible for (a) procurement oflimestone, (b) processing the flue gas and returning it to the stack, (c) delivery ofgypsum to the wallboard manufacturer, and (d) treatment of the wastewater.

II.C Project Objectives/Results

The goal of this project was to demonstrate AFGD retrofit technology forreducing SO2 emissions from coal-fired utility boilers. The project was designed toconfirm pilot plant results and to develop scale-up procedures necessary forcommercial application of the technology, as well as to resolve those technicalissues that could not be adequately addressed in an engineering study or in pilot-scale tests. Specific objectives were to

& Remove at least 90-percent SO2, with a target of 95 percent.& Reduce process cost by one-half that of conventional flue gas

desulfurization (FGD).& Reduce space requirements.& Produce wallboard-grade gypsum.

12

The SO2 removal target was exceeded. SO2 removal efficiency during the testprogram averaged 94 percent, with a maximum of over 98 percent. Thedemonstration facility was operated for about 26,300 hours, with systemavailability of 99.5 percent.

Economic calculations (discussed in section IV.B) show a levelized cost for AFGDat about 63 percent of that for a conventional wet limestone FGD process. This isa significant cost reduction, and is in the range of the target value of 50 percent.The space requirements for AFGD are not discussed in detail in the documentationfor this project, but they are substantially lower than those for conventional FGDprocesses because of smaller equipment and the minimum use of spares. U.S.Gypsum purchased the entire output of 210,000 tons of wallboard-grade gypsum,which had an average purity of 97.2 percent.

II.D Environmental Performance

The AFGD demonstration project showed the capability of AFGD for reducing SO2 emissions at coal-burning power plants. With a coal sulfur content of 3.0percent and a 94-percent SO2 removal rate, the facility reduces emissions by about68,000 tons/yr. The flue gas SO2 content is 0.48 lb/million Btu (MBtu) , which isconsiderably lower than the NSPS of 0.6 lb/MBtu.

With the AFGD unit installed, the Bailly Station became the first power plantamong CAAA Phase I affected units to meet the SO2 standards using FGDtechnology. Recovering wallboard-grade gypsum as a by-product eliminates theneed for solid waste disposal. In addition, use of the WES concept results in zerodischarge of liquids from the plant.

II.E Post-Demonstration Achievements

The AFGD unit at Bailly Station remains in service and will continue to operate forthe balance of the 20-year contract life. As a result of the success of this project, anew venture was formed in 1994 between Pure Air of Manatee, L.P., and FloridaPower and Light Company to provide 1600 MWe of scrubbing capability at itsManatee Power Plant, using the AFGD process and the same own-and-operateconcept as at Bailly Station. The intent was to burn a fuel imported fromVenezuela; Orimulsion is an emulsified bitumen material similar to heavy fuel oilbut having a sulfur content of 3 percent or more, comparable to coal. The Manateeproject was abandoned because of a ban on the use of Orimulsion in Florida.

The project received two major awards: the Outstanding Engineering Achievementaward, from the National Society of Professional Engineers in 1992; and thePowerplant of the Year award from Power magazine in 1993, for demonstratingadvanced limestone FGD technology with innovations in wastewater treatment andgypsum production.

13

III Operating Capabilities Demonstrated

III.A Size of Unit Demonstrated

The demonstration project was conducted at NIPSCO’s Bailly Generating Station,which has two coal-fired boilers. Unit 7, permitted at 183 MWe (gross), beganoperation in 1962, and unit 8, permitted at 345 MWe (gross), began operation in1968. The combined flue gas, representing 528 MWe, is treated in the AFGDscrubber. The combined nameplate rating of units 7 and 8 is 616 MWe (gross), andthe scrubber was designed to accommodate the higher capacity.

Over the three-year demonstration period, the boilers were fired with fiveMidwestern bituminous coals having a sulfur content ranging from 2.21 to 4.73percent. Coal properties are given in Table 1.

Table 1. Coal Properties(Coal Source: Midwestern bituminous; Ultimate Analysis, wt%)

Coal No. I II III IV V

Carbon 66.77 61.46 62.03 59.02 69.39

Hydrogen 4.51 4.38 4.09 4.36 4.94

Nitrogen 1.44 1.23 1.22 1.26 1.17

Sulfur 2.21 2.90 3.21 3.78 4.73

Oxygen 6.73 7.43 8.18 7.18 5.64

Chlorine 0.14 0.10 0.06 0.03 0.07

Moisture 8.63 12.89 11.12 13.69 4.74

Ash 9.57 9.61 10.09 10.68 9.32

Total 100.00 100.00 100.00 100.00 100.00

HHV, Btu/lb 11,932 11,022 10,874 11,000 12,700

III.B Performance Level Demonstrated

The AFGD unit at Bailly Station achieved an average SO2 emissions reduction of 94 percent, with a maximum of over 98 percent. Availability was 99.5 percent.Over 210,000 tons of gypsum, having a purity of 97.2 percent, were produced andsold to a local wallboard manufacturer.

14

80

85

90

95

100

50 60 70 80 90 100

Absorber Recircula tion R ate(Percent o f Design)

SO

Re

mo

val E

ffic

ien

cy

(Pe

rce

nt)

2

S toichiome tric R atio - 1 .045

2.252.754.04.5

S ulfur conten t (% )

2K-1069-3 C

To achieve these results, some modifications to the initial design were required,primarily in materials of construction. At the wet/dry interface within the absorbervessel, it was necessary to install a C-276 alloy cladding over carbon steel. Thehigh pressure nozzles in the original WES were replaced with two-fluid nozzleswhich provided better droplet size distribution, eliminating the problem ofexcessive accumulation of solids in the ductwork.

III.C Major Operating and Design Variables Studied

The variables studied in the test program were (a) sulfur content of the coal, (b)slurry recirculation rate, (c) stoichiometric ratio (SR) of calcium sorbent to sulfurremoved, and (d) liquid to gas (L/G) ratio in the absorber. The results aresummarized below.

& Effects of Recirculation Rate and Coal Sulfur Content: At a constantSR and coal sulfur content, SO2 removal efficiency increases withincreasing slurry flow rate. For example, at an SR of 1.045 and a coalsulfur content of 2.25 percent, SO2 removal efficiency increases from 90percent at a recirculation rate of 50 percent of design to about 97 percentat 100 percent of design. SO2 removal efficiency was found to decreasewith increasing coal sulfur content. It would be expected that SO2 removalefficiency would be greater at higher coal sulfur content because of thegreater driving force. Since this was not the case, an alternative explanationmight be that residence time is limiting. The Final Report (Pure Air 1996)does not address this issue. The results are shown in Figure 3.

Figure 3. Effect of Slurry Circulation on SO2 Removal Efficiency(100 % Boiler Load)

15

80

85

90

95

100

SO

Re

mo

val E

ffic

ien

cy (

Pe

rce

nt)

2

Liq u id to Ga s R a tio - 76% o f De sign

2.252.754.04.5

Sulfur con tent (% )

S to ic h iom e tr ic R atio(M o les C a lc iu m /M o le S O R e m o ve d)

2

1.010 1.025 1.040 1.055 1.070 1.1001.085

2K -106 9 -4 C

80

85

90

95

100

1.02 1 .03 1 .04 1 .05 1 .06 1 .08

SO

Re

mo

va

l Eff

icie

nc

y (

Pe

rce

nt)

2

O p tim u m C o al(3% S ulfu r)

849396

D e s ign L/G (% )

Sto ich iom e tric Ratio(M oles C alc ium /M ole S O R em oved)

2

1 .07

2K-1069-5 C

& Effects of Stoichiometric Ratio and L/G Ratio: At a constant L/G ratioand coal sulfur content, SO2 removal increases with increasing SR. Forexample, at an L/G ratio of 76 percent of design, SO2 removal efficiencyincreases from about 91 percent at an SR of 1.01 to about 98 percent at anSR of 1.10. These results are shown in Figure 4. Similar patterns exist forother L/G ratios, as shown in Figure 5.

Figure 4. Effect of Stoichiometric Ratio on SO2 Removal Efficiency(100% Boiler Load)

Figure 5. Effect of Liquid/Gas Ratio on SO2 Removal Efficiency(100% Boiler Load)

16

III.D Boiler Impacts

Operation of the Pure Air AFGD unit, which treats flue gas downstream of theboiler, had no effect on boiler performance. As indicated above, use of the AFGDprocess brought the generating station into compliance with air pollutionregulations.

III.E Commercialization of the Technology

III.E.1 Current Status

The AFGD unit at Bailly Station will continue to operate for the remainder of the20-year own-and-operate contract, with Pure Air as the owner of the unit and AirProducts as the operator. This facility will reduce SO2 emissions by approximately68,000 tons/yr. The gypsum by-product will continue to be sold to United StatesGypsum Company for manufacture of wallboard.

In April 1994, Pure Air of Manatee, L.P., entered into a contract to provide 1600MWe of SO2 scrubbing capability at Florida Power & Light Company’s Manateepower plant on the same own-and-operate basis. Although the project did not goforward because of the inability to obtain approval for burning Orimulsion inFlorida, the design for the Manatee scrubber features two 800-MWe absorbervessels, PowerChip® gypsum production, and use of the WES concept.

III.E.2 Future Work

In conventional wet limestone scrubbers, dibasic acids such as adipic acid havebeen used as reagent additives to enhance SO2 removal performance. Pure Air isevaluating the possible use of dibasic acids in the AFGD process, taking intoaccount effects on capital and operating costs as well as properties of the by-product gypsum.

Initially, WES operation experienced problems resulting from plugging of the highpressure nozzles. These problems have been solved by use of dual fluid nozzles.Because of ample capacity for wastewater treatment at Bailly Station, the WES isnot currently being operated, but the concept was demonstrated to be viable in thisproject and it most likely would be used in future installations.

Likewise, problems associated with gypsum recovery have been solved, permittingattainment of the goal of zero discharge of waste streams. The PowerChip®system was demonstrated successfully, and is available for use in otherapplications. Since there is added cost associated with running this system, and

17

since United States Gypsum is capable of handling the normal production ofgypsum from the Bailly Station, the PowerChip® system is not being operated ona routine basis.

A brown plume has occasionally been observed emanating from the combined unit7 and unit 8 stack. Studies have identified small amounts of sulfur trioxode (SO3)in the flue gas, generated by oxidation of SO2, as the cause of this phenomenon.Since SO3 is not removed in the scrubber, it is necessary to control this pollutantby other methods; one commonly used approach is injection of ammonia into thestack gas. At Bailly Station, NIPSCO has successfully minimized the concentrationof SO3 in the flue gas by careful control of boiler variables, thus eliminating theneed for ammonia injection.

18

IV Market Analysis

IV.A Potential Markets

The AFGD process is potentially applicable to all types of conventional coal-firedboilers including stoker, cyclone, wall-fired and tangentially fired boilers. Pure Airstates that the most likely application of AFGD is with boilers above 100 MWe incapacity, since duct scrubbers are favored at lower capacities because of lowercapital cost. Of the larger boilers, the most likely retrofit candidates would bethose currently burning medium to high sulfur coal that choose to install scrubbersrather than switch fuels.

In the United States, relatively few of the plants regulated under Phase I of the1990 CAAA have installed scrubbers for SO2 control. A large proportion of theseplants have achieved compliance by fuel switching or by purchasing SO2 emissioncredits. However, the price of emission allowances is increasing, and it is likelythat a significant U.S. market for flue gas desulfurization will develop in Phase II.High performance, cost-effective scrubbers will be prime contenders for a share ofthis market. The international market represents additional opportunities forAFGD.

For the most effective use of all of the features of AFGD, it is essential to identifymarkets for the gypsum by-product. The Pure Air project has shown thatWallboard manufacture provides the perfect use for synthetic gypsum. Based inpart on the success of this project, several new plants are being planned or builtto manufacture wallboard from by-product gypsum. In fact, because of its uniformproperties, synthetic gypsum has displaced the natural material as the preferredfeedstock for wallboard manufacture. Coupled with a shortage of natural gypsum,the market prospects for FGD by-product gypsum appear strong. Another use for FGD by-product gypsum is to combine it with boiler fly ash to make cement.However, cement users tend to be smaller and more numerous, and gypsumspecifications for cement are slightly different than for those for wallboard.

As mentioned previously, Pure Air envisions a new market opportunity for AFGDin East Coast power plants involving the use of a low cost, high-sulfur fuel knownas Orimulsion. If use of this fuel is approved by state regulatory agencies,desulfurization of the flue gas will be required, and the AFGD process will be acandidate for this purpose.

IV.B Economic Assessment of Utility Boiler Applications

IV.B.1 AFGD CostsThe Final Report (Pure Air 1996) includes an economic estimate for a 500-MWeAFGD unit, using a projected process design for the nth plant which incorporates

19

improvements based on experience gained from the demonstration project. Thecapital cost includes a 10-percent retrofit allowance, representing moderate retrofitdifficulty. The coal feed is assumed to contain 3 wt% sulfur, with 90-percent SO2

emissions reduction using an SR of 1.04.

The estimated capital cost is $111/kW in 1996 dollars. For a 15-year project life,the levelized cost on a current dollar basis is 5.3 mills/kWh. This is equivalent to$245/ton of SO2 removed. On a constant dollar basis, the levelized cost is 4.1mills/kWh, equivalent to $188/ton of SO2 removed. These economics are given inmore detail in Table 2.

Since the design assumes zero discharge, there is no waste disposal expense. Theeconomics include a credit of $1.00/ton for the by-product gypsum. Theeconomics also assume a credit of $150/ton for 7379 tons/yr of SO2 emissionallowances. The allowance price appears realistic in the light of current trends, butthe Pure Air Final Report does not explain the basis for selecting the amount ofSO2 allowances sold. Credits significantly effect the total levelized cost, andcontribute substantially to the favorable economics. The issue of allowancequantities and prices needs to be explored in future evaluations of the AFGDprocess.

IV.B.2 Comparison With Other Technologies

The Pure Air Final Report discusses the relative merits of wet and dry scrubbingfor SO2 removal. An advantage of AFGD is its capability of achieving 95-percentSO2 removal at reasonable cost. Pure Air provides some quantitative costcomparisons between AFGD and conventional FGD, assuming a 4.3-percent sulfurfeed and a 30-year project life (as opposed to the 3.0-percent sulfur feed and 15-year project life used in the economic estimate discussed in section IV.B.1, forAFGD alone). Several scenarios are considered, involving a range of plantcapacities and assumptions regarding by-product credits and SO2 emissionallowances.

At a capacity of 500 MWe, the most favorable case for AFGD shows a levelizedcost of $236/ton of SO2 removed compared with $373/ton for a typicalconventional wet process FGD. Thus FGD costs about 58 percent more thanAFGD, or, in other words, the AFGD cost is about 63 percent of the conventionalprocess cost. This represents a significant advantage for AFGD and approaches thestated target for AFGD of about one-half the cost of conventional processes.

Insufficient detail is given in the Pure Air report to permit further analysis of thesefigures. It should be noted that the economic calculations in this comparisoninclude a price of $300/ton for SO2 emission allowances, which is twice thatassumed in the base case. Since emissions credits have a significant effect on theeconomics, these results must be treated with caution.

20

Table 2. Summary of Performance and Cost Data (1996 dollars)

Coal Properties Higher heating value(HHV) 12,500 Btu/lb

Power Plant Attributes With Controls Plant capacity, net 500 MWe

Power produced, net 2.85x109 kWh/yr

Capacity factor 65 %

Coal fed 1.14x106 tons/yr

SO2 Emissions Control Data Removal efficiency 90 %

Emissions without controls 4.8 lb/MBtu

Emissions with controls .48 lb/MBtu

SO2 removed 61,495 tons/yr

Gypsum produced 174,450 tons/yr

Emissions allowances sold 7,379 tons/yr

Total Capital Requirement 111 $/kW

Levelized Cost, Current $ Levelization Factor a mills/kWh $/ton SO2 Removed

Capital charge 0.160 3.06 142

Fixed O&M 1.314 1.16 54

Variable O&M 1.314 1.67 78

Less: Credit for gypsum sales b 1.314 -0.08 - 4

Less: SO2 emission allowances c 1.314 -0.54 -25

Total 5.27 245

Levelized Cost, Constant $ Levelization Factor a mills/kWh $/ton SO2 Removed

Capital charge 0.124 2.37 110

Fixed O&M 1.000 0.89 41

Variable O&M 1.000 1.27 59

Less: Credit for gypsum sales b 1.000 -0.06 -3

Less: SO2 emission allowances c 1.000 -0.41 -19

Total 4.06 188

a Levelization based on 15-year project life, 38% tax rate, 4% inflation, and the following capitalstructure: 50% debt @ 8.5% return, 15% preferred stock @ 7.0% return, and 35% commonstock @ 7.5% return, giving a weighted cost of capital of 7.925% (including inflation).

b $1.00/tonc $150/ton

21

V Conclusions

The AFGD process as demonstrated by Pure Air at the Bailly Station offers a reliable andcost-effective means of achieving a high degree of SO2 emissions reduction when burninghigh-sulfur coals. Many innovative features have been successfully incorporated in thisprocess, and it is ready for widespread commercial use. The system uses a single-loopcocurrent scrubbing process with in-situ oxidation to produce wallboard-grade gypsuminstead of wet sludge. A novel wastewater evaporation system minimizes effluents. Theadvanced scrubbing process uses a common absorber to serve multiple boilers, therebysaving on capital through economies of scale.

Major results of the project are summarized below.

& SO2 removal of over 94 percent was achieved over the three-year demonstration period, with a system availability exceeding 99.5 percent.

& A large, single absorber handled the combined flue gas of boilers generating 528 MWe of power, and no spares were required.

& Direct injection of pulverized limestone into the absorber was successful.

& Wastewater evaporation eliminated the need for liquid waste disposal.

& The gypsum by-product was used directly for wallboard manufacture, eliminating the need to dispose of waste sludge.

22

Abbreviations

AFGD Advanced Flue Gas DesulfurizationARS air rotary spargerCAAA Clean Air Act AmendmentsCaCO3 limestoneCaSO3 calcium sulfiteCaSO4 calcium sulfate (gypsum)CCT Clean Coal TechnologyDOE U.S. Department of EnergyESP electrostatic precipitatorFGD flue gas desulfurizationH2SO3 sulfurous acidH2SO4 sulfuric acidL/G liquid to gasNIPSCO Northern Indiana Public Service CompanyNOX nitrogen oxidesNSPS New Source Performance StandardsSO2

sulfur dioxide

SO3 sulfur trioxodeSR stoichiometric ratioWES wastewater evaporation system

23

References

U.S. Department of Energy. 1989. Comprehensive Report to Congress, Clean CoalTechnology Program—Advanced Flue Gas Desulfurization (AFGD) Demonstration Pro-ject, proposed by Pure Air.

Pure Air. 1996. Final Report, Volume 2—Project Performance and Economics.

24

Bibliography

Ashline, P.M. 1993. “A Case Study: The Commercial Deployment of Pure Air’s Clean CoalTechnology,” Pure Air. Paper presented at Second Annual Clean Coal Technology Con-ference, Atlanta, Georgia, September 1993.

Henderson, J., D.C. Vymazal, D.A. Stryf, and T.A. Sarkus. 1994. “Two Years ofOutstanding AFGD Performance, Pure Air on the Lake’s Bailly Scrubber Facility,” PureAir, and Northern Indiana Public Service Company, and U.S. Department of Energy. Paperpresented at Third Annual Clean Coal Technology Conference, Chicago, Illinois, September1994.

Keeth, R.J., P.J. Ireland, and P.T. Radcliffe. 1991. “Economic Evaluation of 28 FGDProcesses,” United Engineers and Constructors, and EPRI. Paper presented at SO2 ControlSymposium, sponsored by EPRI, EPA, and DOE, Washington, D.C., December 1991.

Manavizadeh, G.B., J.J. Lewnard, D.A. Stryf, and T.A. Sarkus. 1995. "Bailly Station AFGDDemonstration Program," Pure Air, Northern Indiana Public Service Company, and U.S.Department of Energy. Paper presented at Fourth Annual Clean Coal Technology Con-ference, Denver, Colorado, September 1995.

Radian Corporation. 1995. Plume Opacity Modeling of NIPSCO Bailly Unit 7 and 8 Stackwith May 1995 Data..

Vymazal, D.C., C.B. Manavizadeh, and D.W. Smith. 1996. “Economic and EnvironmentalBenefits of Advanced Flue-Gas Desulfurization Technology—Three Years of DOE TestResults,” Pure Air. Paper presented at American Power Conference, Chicago, Illinois, April1996.

Wrobel, B., and D.C. Vymazal. 1992 and 1993. “Acid Rain Compliance—Advanced Co-Current Wet FGD Design for the Bailly Station,” Northern Indiana Public ServiceCompany, and Pure Air. Paper presented at First Annual Clean Coal Technology Con-ference, Cleveland, Ohio, September 1992, and Air and Waste Management AssociationAnnual Meeting, Denver, Colorado, June 1993.